MAR WILEJ RIGHT INTERACTIVE

Top of text

Top of CT
Make My Memory: How
Advertising Can Change
Our Memories of the Past
Kathryn A. Braun
Harvard Business School
Rhiannon Ellis
University of Pittsburgh
Elizabeth F. Loftus
University of Washington

ABSTRACT

Marketers use autobiographical advertising as a means to create

nostalgia for their products. This research explores whether such
referencing can cause people to believe that they had experiences as
children that are mentioned in the ads. In Experiment 1,
participants viewed an ad for Disney that suggested that they shook
hands with Mickey Mouse as a child. Relative to controls, the ad
increased their confidence that they personally had shaken hands
with Mickey as a child at a Disney resort. The increased confidence
could be due to a revival of a true memory or the creation of a new,
false one. In Experiment 2, participants viewed an ad for Disney that
suggested that they shook hands with an impossible character (e.g.,
Bugs Bunny). Again, relative to controls, the ad increased confidence
that they personally had shaken hands with the impossible
character as a child at a Disney resort. The increased confidence is
consistent with the notion that autobiographical referencing can
lead to the creation of false or distorted memory. 䉷 2002 John Wiley
& Sons, Inc. Base of text

Psychology & Marketing

䉷 2002 John Wiley & Sons, Inc. Vol. 19(1):1–23 (January 2002)
1 Base of DF
MAR WILEJ LEFT INTERACTIVE

Top of text
Advertisers play off consumers’ memories and emotions through the use Base of text
of autobiographical referencing. The use of such referencing can cause
consumers to focus less on rational product information and more on
the feelings evoked by their recollected memories (Sujan, Bettman, &
Baumgartner, 1993). Increasingly, marketers are using this technique
to appeal to Baby Boomers where these past images represent lasting
expressions of freedom and youth (Marconi, 1996). For example, Walt
Disney celebrated the 25th anniversary of Disney World in Orlando with
an advertising campaign entitled “Remember the Magic.” The ads re-
sembled vintage home movies and featured scenes of people swimming,
meeting Mickey Mouse, and enjoying themselves on the theme park’s
exciting rides. The campaign’s aim may have been to remind consumers
of their own past happy childhood memories of the park in order to get
them to revisit.
But what if such referencing could change what consumers remember
about their childhood memories of visiting the park? Not all consumers
have had happy experiences at Disney nor do they all have the ability
to accurately conjure up those childhood images at will. Because con-
sumers may use the advertising as a cue to recollect their past ex-
perience, there is the possibility that these recently generated advertis-
ing images may alter what consumers ultimately remember about their
own childhood. After all, there is evidence that cues that get people to
think over and over again about manufactured childhood events can be
a relatively easy way to create false memories or beliefs about child-
hood (Loftus, 1997). Such findings have raised concerns about the ac-
curacy of memories surfaced in hypnosis, guided imagery, or other
prompts in psychotherapy (Lindsay & Read, 1994; Loftus & Ketcham,
1994).
Marketers have found that autobiographical memories may be spon-
taneously activated within the context of an advertising message
(Baumgartner, Sujan, & Bettman, 1992; Krugman, 1967). Marketers
have further shown that they can increase the likelihood consumers will
activate their memories by focusing on experiential information (Wells,
1986) or using dramatic narratives in their advertising campaigns
(Boller, 1990). Autobiographical ads may cause consumers to imagine
themselves in the advertised event, and this vicarious experience may
alter how consumers remember their own past.
What if Disney’s “Remember the Magic” campaign implanted mem-
ories into consumers of things that never happened? The possibility that
marketing stimuli can direct, guide, or change consumers’ autobio-
graphical memories has gone largely untested. This research investi-
gates whether the use of autobiographical referencing can cause ima-
ginings of experiences (even impossible ones) that lead consumers to
become more confident that certain events had happened to them as
children. This possibility holds both managerial opportunities and eth- short
ical ramifications. standard

2 BRAUN, ELLIS, AND LOFTUS Base of RF

MAR WILEJ RIGHT INTERACTIVE

Top of text
Autobiographical Memory Base of text
Autobiographical memory can be defined as memory of past personal
experiences. There has been much attention toward finding ways of ac-
cessing this type of knowledge because it is an important foundation of
one’s self-concept (Hyman, Husband, & Billings, 1994). Sigmund Freud,
for instance, believed his patients had repressed traumatic childhood
memories in their subconscious and believed it was necessary to under-
stand his patients’ original childhood experiences in order to understand
their adult problems.
Psychoanalysts believe childhood is important for understanding re-
lationships because this is when attachment occurs, and those early
relationships are thought to be prototypical of later relationships (Ains-
worth, 1985). As applied to the consumer setting, early childhood brand
relationships may set the emotional stage for later adult brand rela-
tionships. Consumers’ memories of brands or brand experiences from
childhood thus may have a great consequence in their decision making,
as they conjure up those past emotional attachments: “These trails of
autobiographical memories — they are perceived as veridical records ac-
companied by strong visual and, hence, vivid reliving of the original
experience — are not only important in themselves, but especially be-
cause they suggest that the original emotions are also likely to be im-
portant components of autobiographical memories” (Baumgartner et al.,
1992, p. 55, italics added).
Both psychoanalysts and marketers use cues to prod people to re-
member their past. In his work Freud eventually came to the conclusion
that his patients were fantasizing much of their childhood experiences
based on his own suggestions. The marketing-research paradigms have
focused on how brands might associate themselves with actual past con-
sumer experiences. Virtually no research has examined memories of
brand experiences, in particular childhood ones, and the manner in
which the advertising influences those recollections. In light of previous
findings on autobiographical referencing an important and yet unex-
plored question arises: Might exposure to an autobiographical ad alter
consumer’s recollection of a past childhood experience or even create a
memory of an experience that never happened? For instance, some
childhood memories may be based more on recurring ads consumers are
exposed to rather than on recollection of actual childhood events. Sim-
ilarly, some consumers may come to believe that they had taken part in
an experience when in fact they had only viewed an ad of the event.
This alteration is possible because of the reconstructive nature of
memory (see Schacter, 1995 for a full review; Braun, 1999, for its ap-
plication in marketing). A consumer’s past is constantly being updated
to fit one’s changing self-knowledge and social contexts (Bruner, 1986;
Neisser & Fivush, 1994; Spence, 1982). The process of rewriting one’s short
history is natural and allows one to adapt to possibilities in the future standard

MAKE MY MEMORY 3 Base of RF

MAR WILEJ LEFT INTERACTIVE

Top of text
(Hyman & Pentland, 1996). As time passes there lies a greater likeli- Base of text
hood that temporally available information will be used to reconstruct,
and perhaps distort, how the experience is remembered (Thompson,
Skowronski, Larsen, & Betz, 1996).

False Memories and Imagination Inflation

Memories that have had time to fade are particularly subject to distor-
tion. For example, Loftus and Pickrell (1995) suggested to adult partic-
ipants that at age five they had been lost in a shopping mall and rescued
by an elderly person. About a quarter of the adults fell sway to this
suggestion. Using a similar procedure, Hyman and Pentland (1996) sug-
gested and had participants imagine having spilled a bowl of punch at
a wedding as a child. About a quarter of adults fell sway to this sugges-
tion, and even more did so when imagination of the experience was
encouraged. The false memories typically incorporated the punch-bowl
incident into a broader account based on accurate personal knowledge.
These studies show that with suggestion and imagination a significant
minority of people can be led to believe that they had experiences that
were manufactured, and many of them elaborated upon those false ex-
periences with idiosyncratically produced details.
The act of imagining oneself having a childhood experience forces
people to create alternatives to reality (if the experience never hap-
pened; Roese, 2000 discusses other effects of counterfactual thinking).
The ease with which these vividly pictured figments of the imagination
come to mind may promote their acceptance as real regardless of their
actual veridicality (Schwarz, 1996). Such imagining might induce
source attribution errors whereby the recently imagined event becomes
confused with the actual past. Researchers find it is particularly difficult
to detect differences between recent imaging and childhood memory. For
instance, Johnson, Foley, Suengas, and Payne (1988) asked participants
to think of actual or imagined personal events from either the recent
past or childhood and then rate them on a number of characteristics.
They found far fewer significant differences between actual and imag-
ined childhood events than for actual and imagined recent events.
A common result of having people imagine an experience is increased
confidence that the event occurred. Garry, Manning, Loftus, and Sher-
man (1996) looked at the relationship between cues asking participants
to imagine an experience and the later reporting of the event happening
to them as a child. In their paradigm they asked participants to rate
the likelihood certain childhood events happened to them on a life-
events inventory (LEI) containing many experiences, for example, get-
ting lost in a shopping mall. Two weeks later, half of the participants
were instructed to imagine themselves as children experiencing several
of these events, including some that had never happened to them. Only short
standard

4 BRAUN, ELLIS, AND LOFTUS Base of RF

MAR WILEJ RIGHT INTERACTIVE

Top of text
participants who performed the imaginative exercise reported substan- Base of text
tial rises in confidence that both actual and illusory incidents had oc-
curred. The researchers called this effect imagination inflation.

Experimental Investigation
Although there is empirical evidence that suggestions can influence
childhood memory, a question arises as to whether it is possible for a
marketer to exert an influence with suggestions in the form of advertis-
ing. The first study investigates whether autobiographical advertising
can prompt consumers to image their childhood experiences so their
memories become more consistent with the images evoked in the ad-
vertising. That first study focuses on a central childhood experience,
visiting Disney World, and specifically shaking hands with Mickey
Mouse. It is known from past research that pictures or images can trig-
ger stronger remembering (Schacter, Johnson, Angell, & Gross, 1997),
and that actions can be of superior value for prompting reconstruction
because they typically form the unique attribute of a specific event (An-
derson & Conway, 1993). Further, it is known that when people recount
their past experiences they begin by visualizing perceptual details and
embellish their memories based upon those details (Belli & Loftus,
1996). For that reason, the target ad incorporated various images from
the park — from Disney’s glistening castle to the mention of the theme
song from “It’s a Small World.” The ad began with “Remember
the Magic” and describes a day in the park from a child’s perspective,
with the culmination being shaking hands with Mickey Mouse. (See
Figure 1).
It is predicted that if the ad is part of the reconstruction process, the
ad elements or images may be likely to appear as part of consumers’
reconstructed memory of their visit, regardless of whether or not the
events had actually happened. In addition, if the ad causes imaging of
the childhood experience, then this imagination process will lead con-
sumers to believe the ad-based experience of shaking hands with Mickey
happened to them as a child — advertising inflation.

EXPERIMENT 1

The purpose of this study was to determine whether autobiographically

focused advertising could directly affect how consumers remember a
prior childhood experience. This is important for marketers because,
although it is known that past experiences can be an important driver
in future purchase, it is not known whether marketers can evoke or alter
memories of childhood experiences through their advertising messages. short
standard

MAKE MY MEMORY 5 Base of RF

MAR WILEJ LEFT INTERACTIVE

Top of text
Base of text

Figure 1. Autobiographical ad.

short
standard

6 BRAUN, ELLIS, AND LOFTUS Base of RF

MAR WILEJ RIGHT INTERACTIVE

Top of text
Method Base of text
Subjects. The participants were 107 undergraduates (64 female, 43
male) from a Midwestern university who received course credit for their
involvement.

Design. A single-factor between-subjects design was used. Half the par-

ticipants received the Disney ad, the other half received a control, non-
Disney ad. The participants were randomly assigned to one of these
conditions. Random assignment left 46 in the Disney ad condition, 51
in the control condition.

Materials Procedure. The experimental procedure was adapted from

the Garry et al. (1996) imagination inflation paradigm. On the first week
participants were given a list of 20 childhood events and asked to in-
dicate whether or not the events had happened to them under the age
of 10. They rated these on a 100-mm line where 0 ⫽ definitely did not
happen, 100 ⫽ definitely did happen. The target item was “Met and
shook hands with a favorite TV character at a theme resort.” This item
appeared fourth on the list. The LEI survey was given within several
other experimental tasks. They were asked to come back the following
week to finish one of those experiments.
The following week half were given the Disney ad, the other half the
control ad, by another experimenter. They were encouraged to imagine
themselves experiencing the situation appearing in the ad, and were
given 5 minutes to write down how the ad made them feel and what it
made them think about. Afterwards, participants rated the ad on four
bipolar attitude scales anchored by: “unfavorable,” – “favorable,” “bad –
good,” “unpleasant – pleasant,” and “negative – positive” on a 100-mm
line, where a higher score indicates more favorable attitudes. They rated
how involving the ad was for them, with the use of empathy measures
adapted from the Wells R scale (1986): “I felt I was right there in the
ad experiencing the situation again,” “I really got involved in the feel-
ings provoked by the ad,” “While I was looking at the ad, I could easily
put myself in the situation,” and “While looking at the ad, I felt that the
events were happening to me.” These measures were anchored from
0 ⫽ “strongly disagree” to 100 ⫽ “strongly agree.”
After a 5-minute distraction task, the original experimenter from
Week 1 came in, appearing to be rather panicked, and said there had
been a problem with coding the autobiographical data and could they
please fill out the life-events inventory again.
After a 15-minute distraction task, participants were asked about
their memories of Disney in a questionnaire delivered by a third exper-
imenter. Participants were asked directly if they had ever visited Disney
before they were 10 and if so, to describe their memory of that event. short
They were asked how well they remembered the event (1 ⫽ “not at all,” standard

MAKE MY MEMORY 7 Base of RF

MAR WILEJ LEFT INTERACTIVE

Top of text
7 ⫽ “perfectly”) whether the memory was pleasant (where ⫺3 ⫽ “ex- Base of text
tremely unpleasant,” 3 ⫽ “extremely pleasant”), the emotional involve-
ment in the experience (1 ⫽ “nothing,” 5 ⫽ “extreme”), centrality to their
childhood (1 ⫽ “not central,” 3 ⫽ “central”), and personal importance to
their childhood (0 ⫽ “trivial,” 100 ⫽ “very important”). These scales
were adapted from Thompson et al. (1996). Last, as a means to assess
demand characteristics, participants were asked what they thought the
purpose of the experiment was, and whether or not they believed their
memories of Disney had been influenced by the advertising.

Results
Coding Procedure. Two independent judges coded participants’ reac-
tions to the ads and the recall statements of their Disney exper-
ience. They had no knowledge of the experimental hypotheses. Their
interrater reliability was 0.83 and an average was used for the analysis.

Autobiographical Effects of the Advertising. The judges looked at

participants’ written protocols to determine whether there was mention
of a past experience, a report that the ad caused the participant to imag-
ine or evoke his or her experience, or mention of future expectations
regarding the resort. Out of the 46 participants who received the auto-
biographical ad, 30 (or 65%) mentioned memories of Disney World, 34
(or 74%) mentioned that the ad caused them to imagine the experience,
and 29 (or 63%) mentioned future visits to the park. For instance, one
participant wrote:

It made me think back to when I was a kid and went to Disney World.
It was great. I remembered eating all day long, riding on Space Moun-
tain for the first time, and especially meeting Mickey Mouse! It made
me want to beg my parents to go there over Christmas break. I want
to re-live the memories I had as a child.

As Baumgartner et al. (1992) found, this reliving can cause nostalgic

feelings:

I felt very nostalgic, remembering what it was like the first time I was
at Disney World. Disney World seems like a fantasyland “dreamlike”
place to visit. A magical place where memories take place.

Even those who had not visited the park in the past were able to
generate this imagination process: “It made me want to visit Disney
World even more than I already want to. It describes it just like I imag-
ine it to be.”

Imagination Inflation. To reiterate, a major goal of Experiment l was

to access whether exposure (plus imagination) to an advertisement short
standard

8 BRAUN, ELLIS, AND LOFTUS Base of RF

MAR WILEJ RIGHT INTERACTIVE

Top of text
about a product can increase participants’ confidence that certain events Base of text
occurred in their own past childhood experience with that product. The
first analysis explored whether participants became more confident that
the critical event occurred by examining whether LEI scores moved up
or down, for the critical item “shaking hands with a TV character at a
theme resort.” The difference between Week 2 and Week 1 was the de-
pendent measure — a score of 0 would indicate no memory change, a
positive score a memory inflation, and a negative score a memory defla-
tion.
Because the interest was in seeing if the ad could make the event
more probable, people who were (at Week 1) quite certain that they had
already experienced the shaking-hands event were eliminated from the
data set. That left those who reported the shaking-hands incident had
a low likelihood of occurrence on the first LEI, defined as a 0 – 50 on the
line mark scale, to determine if the ad increased their confidence the
event had occurred (as per Garry et al., 1996), which left 73 participants
(32 control, 41 Disney). There are several ways to analyze this data. The
first analysis looked at the number of people in each group that showed
an increase, decrease, or same report on the LEI (see Figure 2). Signif-
icantly more people who received the Disney ad went up on the LEI,
90% versus 47% in the control, significantly different at ␹2(2, N ⫽ 73)
⫽ 17.3, p ⬍ .0001. In addition, mean difference scores on the LEI were
analyzed. A positive difference would indicate greater confidence the
event had happened to them, a negative difference less confidence. The
Disney group showed a more positive change, M ⫽ 37.05, than the

Figure 2. Experiment 1: Change in LEI “shaking hands.” short

standard

MAKE MY MEMORY 9 Base of RF

MAR WILEJ LEFT INTERACTIVE

Top of text
control group, M ⫽ ⫺ 1.5 significantly different at t(71) ⫽ 5.93, p ⬍ Base of text
.0001. There were no significant differences on the other 18 items on
the LEI test.
Another question of interest is which (if any) advertising measures
would best predict the advertising/imagination inflation. Earlier work
by Krugman (1967) suggests that autobiographical referencing may in-
crease involvement, and Wells (1986) argues that traditional attitude-
toward-ad measures may not effectively capture this effect. The four
traditional ad-rating measures, where participants indicated their over-
all liking for the ad, loaded on one factor with coefficient alpha ⫽ .92
and were combined into an ad attitude/rating index. The Wells R in-
volvement with ad measures, where participants indicated how emo-
tionally involved they became in the ad-based situation, also loaded on
one factor with coefficient of 0.90 and were combined to form an involve-
ment index. Here it was found that the correlation between the ad at-
titude index and the LEI memory change was r ⫽ .08 and involvement
in ad measures was r ⫽ .31, showing that in absence of the LEI change
data, the Wells scale can capture the memory effects slightly better than
traditional ad-attitude measures can.

Disney Memory. Thirty-four participants from the Disney ad group and

34 from the control group reported having visited a Disney park in the
past. They were asked to describe and rate that visit on several scales.
The two judges coded the recall statements on the number of words,
independent thoughts, positive thoughts, negative thoughts, and ele-
ments from the ad that were mentioned. For the latter, the judges
looked at the recall statements for usage of words from the ad (e.g.,
“magical”) and actions from the ads (e.g. “cool rides” or hearing “It’s a
Small World”). One participant, who appeared to be heavily influenced
by the advertising, wrote:

Went on lots of different rides, met Mickey Mouse, watched the “Small
World” show . . . hug . . . Got sick from too much junk food, got
drenched on log ride . . .

The ad had featured meeting Mickey Mouse, the “It’s A Small World”
song, and a variety of rides and food. That response would have scored
a 4 on elements. Because some of these ad-suggested ideas may natu-
rally appear in one’s recall of Disney, the autobiographical ad condition
was compared to the control condition to assess whether the ad in-
creased their appearance. There were no significant differences in the
length or number of thoughts between the Disney/control condition;
M ⫽ 31 words in the Disney condition, M ⫽ 26 in the control condition;
M ⫽ 4.4 thoughts in the Disney condition, M ⫽ 4.1 in the control con-
dition. There were significantly more positive thoughts (M ⫽ 3.6) in the short
Disney ad condition than in the control (M ⫽ 2.8, t[66] ⫽ 2.07, p ⫽ .02). standard

10 BRAUN, ELLIS, AND LOFTUS Base of RF

MAR WILEJ RIGHT INTERACTIVE

Top of text
There were no differences in negative thoughts; M ⫽ 0.79 in Disney, Base of text
M ⫽ 0.50 in control, t ⬍ 1. There were more ad elements appearing in
the recall in the ad condition (M ⫽ 2.38) than in the control (M ⫽ 1.47,
t[66] ⫽ 2.65, p ⫽ .005).
Participants also rated their memory on several objective measures;
means appear in Table 1. Those who received the Disney ad rated their
memories as being more personally important, t(66) ⫽ 1.93, p ⫽ .025;
better remembered, t(66) ⫽ 3.11, p ⫽ .001; more pleasant, t(66) ⫽ 2.17,
p ⫽ .015; and more central to their childhood, t(66) ⫽ 1.59, p ⫽ .055
than those who did not receive the Disney ad. There were no significant
differences in emotive aspects of recall, t ⬍ 1.

Demand. Participants were asked to guess the purpose of the experi-

ments in order to assess demand factors. No one correctly responded
that the advertising was intended to alter their childhood memories.
Additionally, participants were asked whether the advertising helped
them remember anything from their own experience that they might

Table 1. Objective Memory Measures

Experiment 1
Disney Control
Personal importance (0 ⫽ trivial, 64 51
100 ⫽ very important)
How well remembered (1 ⫽ not 4.8 3.9
at all, 7 ⫽ perfectly)
Pleasantness of visit (⫺3 ⫽ ex- 2.4 2.0
tremely unpleasant, 3 ⫽ ex-
tremely pleasant)
Centrality to childhood (1 ⫽ not 2.2 1.7
central, 3 ⫽ central)
Emotional involvement (1 ⫽ 3.6 3.5
nothing, 5 ⫽ extreme)
Experiment 2
Ariel Bugs Nonautobiographical
Personal importance (0 ⫽ trivial, 5.6 5.5 5.0
10 ⫽ very important)
How well remembered (1 ⫽ not 4.4 4.3 3.9
at all, 7 ⫽ perfectly)
Pleasantness of visit (⫺3 ⫽ ex- 2.1 2.4 2.1
tremely unpleasant, 3 ⫽ ex-
tremely pleasant)
Centrality to childhood (1 ⫽ not 2.3 2.2 1.8
central, 3 ⫽ central)
Emotional involvement (1 ⫽ 3.8 3.8 3.5 short
nothing, 5 ⫽ extreme)
standard

MAKE MY MEMORY 11 Base of RF

MAR WILEJ LEFT INTERACTIVE

Top of text
not otherwise have been able to remember. Forty-one participants who Base of text
received the Disney ad said “no”; 5 said “yes.” They were also asked if
the ad they saw for Disney made them remember their experience to be
different than it really was. For instance, did the ad make them remem-
ber having a positive experience when in fact it was very negative?
Forty-three participants said “no,” and only 3 said “yes.”

Discussion
The purpose of this study was to determine whether an autobiographical
ad could make consumers more confident that they had experienced an
advertised-suggested event as a child. It was found that autobiograph-
ical advertising can indeed induce this effect. The participants became
more confident that they had shaken hands with a TV character after
viewing the advertising than if they had not received that retrieval cue.
Figure 3 depicts three routes that could be happening as participants
recalled the shaking-hands event, adapted from Haber and Haber
(1996). Route 1 represents those who rated the shaking-hands experi-
ence high on both the LEI pre- and posttests; in other words, they did
not need the ad to retrieve that memory, and those people were not
included in the LEI analysis for that reason. Routes 2 and 3 are the
most interesting with respect to the role autobiographical advertising
has on cuing past experiences, for both routes represent those individ-
uals that increased on the LEI test after receiving the Disney ad. The
paradox of offering a retrieval cue is that it can help access both true
and false memories (Spiegel, 1997). As Garry et al. (1996) note in their
study of imagination inflation, some of the observed increases could be
due to the ad providing an effective cue in surfacing a veridical memory.
In other words, perhaps some participants at Week 1 forgot about shak-
ing hands with Mickey and the ad helped surface that encounter. How-
ever, it is possible that the ad may have implanted that experience in
consumers’ minds.
It was impossible to determine whether in fact participants had ever
shaken hands with Mickey Mouse, and thus is not clear whether Routes
2 or 3 were occurring. One way to determine this would be to get par-
ticipants to remember an event that would have been impossible, for
example, getting them to remember shaking hands with a character
who became popular later, but did not exist at the time the participant
would have been to a Disney park or, alternatively, feature a competi-
tor’s character (e.g., Warner Bros. Bugs Bunny), who would have never
been in the park. If advertising leads people to claim an impossible ex-
perience, there would be evidence for the creation of false memories.
This reasoning guided the design of Experiment 2. short
standard

12 BRAUN, ELLIS, AND LOFTUS Base of RF

MAR WILEJ RIGHT INTERACTIVE

Top of text
Base of text

Figure 3. Routes to memory retrieval.

EXPERIMENT 2

The purpose of this experiment was to determine whether false infor-

mation in advertising about childhood experiences at Disney could make
consumers believe those events had happened to them. Two types of
false information were tested within the same ad format as used in
Experiment 1: In one ad it was suggested that they had shaken hands
with Bugs Bunny (a Warner Bros. character); in another ad it was sug-
gested they had shaken hands with Ariel, the Little Mermaid (a Disney
character not yet introduced at the critical childhood time period, before
the age of 10). Bugs, meanwhile, was around during participants’ child- short
standard

MAKE MY MEMORY 13 Base of RF

MAR WILEJ LEFT INTERACTIVE

Top of text
hood but would have never have been at a Disney resort. A nonautobio- Base of text
graphical Disney ad served as a control group in order to investigate
whether it is the autobiographical nature of the ad that is influencing
consumers’ past or the mere Disney mention that is triggering the re-
sults.1 The nonautobiographical ad was more informational in nature
(Puto & Wells, 1983), focusing on a new ride being offered at the park
and providing information about how to order park tickets via the Web
at a discount price.

Method
Subjects. The participants were 167 (104 female, 63 male) psychology
undergraduates at a western university who received credit for their
involvement.2

Materials and Procedure. The same basic procedure from Experi-

ment 1 was followed, with some adaptations. Participants took the first
LEI during a mass-testing situation and indicated whether or not they
would be interested in participating in an upcoming ad study. Those
who said “yes” were contacted and were assigned to one of the three ad
conditions. Some screening occurred: Those who had visited Six Flags
were eliminated to reduce a possible source-confusion error (Bugs
Bunny does make an appearance at that park).
The target LEI measure was modified to “shaking hands with a car-
toon character in a theme park” with a 10-point scale anchored by “def-
initely did not happen”/“definitely did happen.” The Disney World focus
was changed to Disneyland because of the higher likelihood that the
students would have visited that California resort. All dependent mea-
sures were converted to numbered scales instead of the 100-mm line
marks used in Experiment 1. The discrete numbering system simplified
analysis.
The advertising measures were also the same as Experiment 1, but
were measured on an 8-point scale, with higher values indicating
greater liking or involvement. A few measures were added to determine
whether in fact the false information had become a part of participants’
childhood memories. The participants were asked to indicate with what
confidence they had seen some characters at the park (with Bugs and
Ariel being the target items buried within a set of other characters) and
then to specifically remember what characters they had actually shaken
hands with (again Bugs and Ariel were buried within a list). Specifically,

1
The researchers felt it was important to have all groups imagine the Disney experience to see
what effect the autobiographical referencing had by just mentioning the Disney name, as other
researchers have found spontaneous activation of autobiographical memories in the presence of
advertising (Baumgartner et al., 1992).
2
Differences in cell sizes in reported tests result from some participants skipping over measures. short
This did not occur frequently nor in an uneven distribution across conditions. standard

14 BRAUN, ELLIS, AND LOFTUS Base of RF

MAR WILEJ RIGHT INTERACTIVE

Top of text
they were asked to indicate their confidence that they had met the fol- Base of text
lowing characters, with Bugs appearing third and Ariel appearing fifth
on a list of eight characters, and to put an “X” in front of characters they
had actually met in person (e.g., shaking their hand); the Bugs and Ariel
placement was the same on this list of eight characters.

Results
Autobiographical Effects of Advertising. The two autobiographical
ads were more involving for participants, scoring higher than the non-
autobiographical Disney ad on the Wells R scale index, formed as in
Experiment 1 except that the scale now ranged from 1 to 8, with lower
scores indicating less involvement. Means follow: 5.2 for the Ariel, 5.1
for the Bugs, 3.8 for the nonautobiographical Disney ad; a t-test com-
parison showed the two autobiographical ads were significantly higher
than the nonautobiographical ad, t(165) ⫽ 4.3, p ⬍ .0001.3
The ad attitude/rating index showed no significant difference across
the groups; the new range for attitude scale was 1 to 8, with lower values
indicating less-favorable attitudes. Means follow: M ⫽ 5.9 Ariel, M ⫽
5.8 Bugs, M ⫽ 5.5 nonautobiographical Disney, F ⬍ 1. Therefore, the
Wells index can better differentiate types of advertising that influence
memories than traditional ad-attitude measures.

Imagination Inflation. Recall that participants indicated how likely

the target event, shaking hands with a cartoon character at a theme
park, happened to them before the age of 10, on a 10-point scale, where
1 ⫽ “definitely did not happen” and 10 ⫽ “definitely did happen.” They
filled out the life-events inventory twice, once during the mass testing,
and then again in a subsequent session. The discrete numbering scale
allowed for a more precise segmentation than Experiment 1. Those who
had indicated the event was unlikely to have happened to them (scoring
1 – 5 on the initial test) were eliminated, which left 106 participants (34
Ariel, 36 Bugs, 34 nonautobiographical ad condition). The interest was
in determining if those individuals who had rated the event as being
relatively unlikely would increase their confidence it had happened to
them after seeing the false autobiographical suggestions in the adver-
tising. Figure 4 depicts those individuals from each group and the di-
rection of movement on the target LEI item, “shaking hands.” All groups
showed a tendency to increase but it was more pronounced in the con-

3
Because the interest was in the difference between autobiographical and nonautobiographical
advertising effects, the two autobiographical groups were combined after finding no significant
difference between them in order to assess if they were different from the other type of ad.
Another way to analyze this would be to run an ANOVA on the three groups and do post-hoc
comparisons. This was done as well, with similar results. For simplicity, only the t-test results
are reported. In these t tests pooled variance was used, and when directionality was predicted, short
one-tailed p values were reported. This technique was used throughout Experiment 2. standard

MAKE MY MEMORY 15 Base of RF

MAR WILEJ LEFT INTERACTIVE

Top of text
Base of text

Figure 4. Experiment 2: Change in LEI “shaking hands.”

ditions that received an autobiographical ad — 76% in the Ariel condi-

tion, 78% in Bugs, and 62% in the nonautobiographical condition — the
autobiographical groups being significantly more likely at ␹2(1, N ⫽ 106)
⫽ 2.5, p ⫽ .05. Mean movement on the LEI was also compared across
groups, M ⫽ 4.8 for Ariel, 5.1 for Bugs, 3.8 for nonautobiographical
Disney. Participants in the autobiographical ad conditions displayed
significantly greater movement in their confidence that they had shaken
hands than their nonautobiographical counterparts; t(104) ⫽ 1.95,
p ⫽ .05.
As found in Experiment 1, the Wells R scale was more correlated with
the change in LEI than the ad attitude scale, r ⫽ 0.2 Wells, r ⫽ ⫺ .1 ad
attitude.

Disney Memory. All participants were asked to write about and rate
their memory of their childhood experience at Disney; 46 in the Ariel
condition, 56 in the Bugs condition, and 46 in the nonautobiographical
condition did so. Those that received the autobiographical ads reported
better clarity in their memories, t(109) ⫽ 2.07, p ⫽ .02; more emotional
content, t(111) ⫽ 1.64, p ⫽ .05; more centrality of memory to childhood,
t(121) ⫽ 2.26, p ⫽ .01; and more importance to their childhood, t(110)
⫽ 1.76, p ⫽ .04. There was no significant difference on pleasurable mem- short
ories between conditions. Means appear in the bottom of Table 1. standard

16 BRAUN, ELLIS, AND LOFTUS Base of RF

MAR WILEJ RIGHT INTERACTIVE

Top of text
False Memory. Participants were asked more directly about specific Base of text
images (false ones) that may have happened on their childhood visit to
the park. The interest was whether they would report greater confidence
in having met or remembering shaking hands with either Ariel or Bugs
if they received those respective ads. They were first asked to indicate
confidence that they had met several characters at the park on a scale
of 1 to 7, anchored by “not very confident” and “very confident.” Confi-
dence in meeting Bugs was M ⫽ 2.7 for the Bugs group, M ⫽ 2.1 for
Ariel, M ⫽ 2.2 for the nonautobiographical Disney ad. A t-test compar-
ison found the Bugs group to be more confident than the other groups
that they had met him, t(114) ⫽ 1.67, p ⫽ .04. Participants were also
asked whether they remembered specifically shaking hands with Bugs
at the park as a child, 16% in the Bugs condition remembered doing so,
compared to 7% in the Ariel group, and 7% in the nonautobiographical
group. A comparison between the Bugs and other-ad conditions was
significant, 16% versus 7%, ␹2(1, N ⫽ 131) ⫽ 2.6, p ⫽ .05. In contrast,
there was no significant difference between the Ariel-exposed group and
other ad conditions on confidence they had met Ariel at the park, M ⫽
1.3 for Ariel, M ⫽ 1.8 for Bugs, M ⫽ 1.3 for nonautobiographical Disney.
However, there was a greater tendency for the subjects to believe they
had shaken hands with Ariel as a child if they received the false ad: 7%
of those in the Ariel condition remembered shaking her hand compared
to 0% in the Bugs condition, and 4% in the nonautobiographical Disney
condition; a comparison between the Ariel and other-ad groups was sig-
nificant at ␹2(1, N ⫽ 131) ⫽ 3.49, p ⫽ .03.

Discussion
The main finding of this experiment is that featuring impossible events
in autobiographical advertising can cause people to believe they had
experienced the events. For example, 16% of people claimed that they
shook hands with Bugs after receiving the false Bugs ad; 7% said they
remembered meeting Ariel, a character that had not yet been intro-
duced, after seeing an ad suggesting all children meet her at the park.
Neither character would have been at the park during the participants’
childhood, demonstrating that autobiographical ads can alter memories
of the past. As found in Experiment 1, imagination inflation occurred,
where people became more confident that the ad-suggested event had
happened to them when they were children.
The Bugs ad was more effective in inducing this memory change. This
could be because Bugs is a better-known character, and thus easier for
participants to imagine during the ad exposure. For instance, in their
study of imagining’s effect on prediction of future actions, Sherman,
Cialdini, Schwartzman, and Reynolds (1985) found that easier-to-imag-
ine illnesses were judged as more likely to occur. Goff and Roediger short
(1998) found that as the number of imaginings of false experiences in- standard

MAKE MY MEMORY 17 Base of RF

MAR WILEJ LEFT INTERACTIVE

Top of text
creased, so did the reported memories of participating in these events. Base of text
Perhaps if participants had received several cues indicating the false
characters were associated with Disney, there would be a higher re-
porting of impossible memories. There is some evidence to suggest this
might be the case: There was a relationship observed between confi-
dence in having met Bugs and memory of the shaking-hands experience,
which was missing in the Ariel condition. This participant seemed to
accept the Bugs ad, despite her initial feeling that it was incorrect:

The first thing that went through my mind when looking at the ad was
“why is Bugs Bunny on a Disneyland ad . . . isn’t he a Warner Bros.
character?” I’m still confused about whether he is or not, but oh
well . . .

The advertising may make an event seem more plausible and help
consumers develop (or amend) a script for the childhood experience, two
factors thought to be important in generating a false memory (Pezdek,
Finger, & Hodge, 1997).

GENERAL DISCUSSION

The two studies provide empirical evidence for autobiographical adver-

tising’s influence on how consumers remember their past. Specifically,
it was found that autobiographically focused advertising can make
events (even impossible ones) seem more likely to have happened to
consumers as children. In Experiment 1 it was found that an autobio-
graphical ad increased consumers’ confidence that they had shaken
hands with Mickey Mouse in a before-age-10 visit to the park. There
was no way to determine whether the ad had activated a true memory
or created a new, false one. Experiment 2 found an increase in confi-
dence that participants had experienced the ad-suggested scenario, but,
in addition, found them more likely to believe they had met characters
at the park who would not have been at Disney during their childhood.
The idea that autobiographical advertising can influence how con-
sumers remember their past is a timely issue. Manufacturers like Oval-
tine娃, Alka Seltzer娃, Maxwell House娃, and Shake-n-Bake娃 have be-
gun to dig into their vaults from the 1950s and 60s to pull out nostalgic
images from past advertising campaigns. Undeniably, such ads tap into
some existing consumer memories from their childhood. Marketers had
believed the process began and ended with the ads cuing actual past
experiences, as Freud’s general belief regarding the special status of the
original memory has lingered.
But times are changing, and some marketers are beginning to realize
that memories are constructive. Some have even benefited from the fact short
that their consumers’ memories have been manufactured. Take, for ex- standard

18 BRAUN, ELLIS, AND LOFTUS Base of RF

MAR WILEJ RIGHT INTERACTIVE

Top of text
ample, Stewart’s root beer. They report many adults seem to remember Base of text
growing up drinking Stewart’s frosty root beer in bottles. This is impos-
sible, because the company only began full-scale distribution 10 years
ago, and prior to that only fountain drinks were available. It could be
that glass bottles adorned with sayings like “original” “old-fashioned”
and “since 1924” provide consumers the illusion of a past that they
might have shared as a child. In fact, the vice president of Stewart’s
marketing swears he remembers drinking their soda after Little League
games in an area where distribution was unlikely, but admits, “Mem-
ories are always better when they’re embellished” (Prince, 2000).
Although there is no direct evidence Disney altered memories
through their “Remember the Magic” campaign, the evidence collected
here suggests it is at least possible. The power of memory alteration is
that consumers are not aware they have been influenced. The feeling
associated with remembering a past event, of “seeing” the event unfold
in their mind’s eye, provides one the belief that how it is recollected is
how it happened. Participants held this belief about the permanence of
memory; most indicated that the ad had little (or no effect) on their
recollections. One participant wrote:

While I was reading it (the Disney ad) I was thinking that I have to
take my children there someday because of the great memories that I
have from there and probably always will have.

Most people are under the belief that memory is a permanent store
and with hypnosis or other special techniques past information can be
recovered (Loftus & Loftus, 1980).
Autobiographical advertising, like hypnosis, allows the consumer to
become personally involved in the message. Consumers, for the most
part, enjoy the trip back in memory, and the marketer benefits from the
positive affect brought forth by their role as memory guide. As a guide,
the marketers’ message has consequences on what consumers ulti-
mately remember. Those who received the advertising were more con-
fident the childhood experiences suggested in the advertising had hap-
pened to them.
Marketers can use autobiographical referencing to re-establish a per-
sonal relationship with their consumers. These autobiographical recol-
lections serve important social and personal functions for consumers.
For example, one participant wrote:

It brought back memories indeed. I guess I thought the story-beginning

was a bit cheesy but then I realize I do have to remember what the
experience was like when I was younger. It made you want to run to a
Walt Disney Resort, I must admit. It sounds like a lot of fun! The theme
“Remember the Magic” I think really caters to all ages, even adults,
who may have forgotten what the “magic” was like — take me, for in- short
stance! standard

MAKE MY MEMORY 19 Base of RF

MAR WILEJ LEFT INTERACTIVE

Top of text
Another discussed the social aspects of this remembering process, Base of text
writing that the ad “reminds them of happy memories and encourages
them to share with others.” Marketers can benefit on two counts by
being the memory guide: directly, by influencing the consumer’s attitude
and purchase intention, and indirectly, by activating positive word-of-
mouth regarding the brand.
Remembering is often a social activity in which people come to some
agreement about the past (Hyman et al., 1995) and it is much more
likely to take place in the context of friends, family, or advertising than
with psychotherapists. In some sense, life is a continual memory alter-
ation experiment where memories continually are shaped by new in-
coming information (Hyman et al., 1995). This brings forth ethical con-
siderations. Is it all right for marketers to knowingly manipulate
consumers’ past? On one hand, the alteration will occur whether or not
that was the intent of the marketer. And, in most cases, the advertiser
is unlikely to try to plant a negative memory, as has been the issue with
false memories of childhood abuse. On the other hand, there are ways
in which the marketer can enhance the likelihood consumer memories
will be consistent with their advertising messages. At the very least,
consumers ought to be aware of that power.
Autobiographical referencing is one way to influence consumers’ rec-
ollections. The ads employed in these studies accomplished that by pro-
viding consumers the imagery tools with which to build their memory.
According to Reconstructionist theorist Frederic Bartlett (1932), an im-
age becomes the device for picking bits out of schemes (of existing knowl-
edge) and organizing it in such a way that decreases the variability in
the reconstruction of the past situation. Typical ad rating scales may
not be effective in determining whether an ad will or will not be suc-
cessful in bringing forth consumer memories. This research found an-
other scale, Wells R, to be slightly better in assessing the potential ela-
borative effects.

LIMITATIONS AND FUTURE RESEARCH

The subject pool was fairly homogeneous undergraduates. Generations

exposed to similar media will become more homogeneous in the types
of memories they recall. Different generations may have distinct time
posts or markers that should be targeted for the Disney experience, for
example, different characters or rides. Holbrook and Schindler (1996)
find that there are critical periods for which particular experiences are
imprinted and learning and affect are created for the brand or service.
Childhood experiences may have been a critical period in the foundation
of consumer’s knowledge about Disney, though for other product cate- short
gories different ages might be better suited for memory revision (e.g., standard

20 BRAUN, ELLIS, AND LOFTUS Base of RF

MAR WILEJ RIGHT INTERACTIVE

Top of text
drinking a particular beer in college). Future research should address Base of text
which life periods should be targeted in autobiographical ads, depending
on consumer age and product category.
The research environment also presents limitations. Participants
were instructed to think about the advertising as it was presented to
them. Such directed attention might not be representative of actual ad-
vertising exposures. Future research might vary the level of advertising
exposure to determine how that influences the recollection process.
There are many opportunities for future investigations into this issue
of memory alteration. For example, one avenue might look at what type
of consumers are most likely to be influenced by autobiographical refer-
encing. Forty-eight percent of the population is susceptible to hypnotic
suggestion; these consumers might also be the ones most apt for brand
memory distortion. The Disassociated Experience Scale (DES) has been
fairly robust in capturing this effect (average r about 0.3).
Because Mickey Mouse is a well-known icon of Disney and is probably
already a central part of consumer participants’ script of visiting the
park, the event of shaking hands may have been easier to imagine than
with a lesser-known character (Ariel) or an unrelated character (Bugs).
Not all participants in Experiment 2 easily accepted the false sugges-
tion, as this participant wrote:

Well first off I notice that Bugs Bunny is a Warner Bros. character
rather than Walt Disney so in essence I feel a sense of false advertising
here from the Disney company.

Previous work on false memories finds that very similar characters,

for example, Mickey Mouse and Minnie, are easily substituted for one
another in memory (Loftus, 1997). The participants had only one ad-
vertising exposure to associate these newer characters with Disney. It
is quite possible that this research underestimates the true effects that
autobiographical advertising has on shaping consumers’ pasts because
in the marketplace consumers may see the ads several times.

REFERENCES

Ainsworth, M. D. (1985, Spring). Patterns of attachment. Clinical Psychologist,

38, 27 – 29.
Anderson, S. J., & Conway, M. A. (1993). Investigating the structure of auto-
biographical memories. Journal of Experimental Psychology: Learning,
Memory, & Cognition, 19, 1178 – 1196.
Bartlett, F. C. (1932). Remembering. Cambridge: Cambridge University Press.
Baumgartner, H., Sujan, M., & Bettman, J. R. (1992). Autobiographical mem-
ories, affect and consumer information processing. Journal of Consumer Psy-
chology, 1, 53 – 82. short
Belli, R. F., & Loftus, E. F. (1996). The pliability of autobiographical memory: standard

MAKE MY MEMORY 21 Base of RF

MAR WILEJ LEFT INTERACTIVE

Top of text
Misinformation and the false memory problem. In D. C. Rubin (Ed.), Remem- Base of text
bering our past: Studies in autobiographical memory (pp. 157 – 179).
Boller, G. W. (1990). The vicissitudes of product experience: “Songs of our con-
suming selves” in drama ads. In M. E. Goldberg, G. J. Gorn, & R. W. Pollay
(Eds.), Advances in consumer research (Vol. 17, pp. 621 – 625).
Braun, K. A. (1999). Postexperience advertising effects on consumer memory.
Journal of Consumer Research, 25, 319 – 334.
Bruner, J. (1986). Actual minds, possible worlds. Cambridge, MA: Harvard
University Press.
Garry, M., Manning, C. G., Loftus, E. F., & Sherman, S. J. (1996). Imagination
inflation: Imagining a childhood event inflates confidence that it occurred.
Psychonomic Bulletin & Review, 3, 208 – 214.
Goff, L. M., & Roediger, H. L. III. (1998). Imagination inflation for action
events: Repeated imaginings lead to illusory recollections. Memory & Cog-
nition, 26, 20 – 33.
Haber, R. N., & Haber, L. (1996). Antecedent conditions and operational defi-
nitions for recovered memory effects. Psychonomics Society Convention, Chi-
cago.
Holbrook, M. B., & Schindler, R. M. (1996). Market segmentation based on age
and attitude toward the past: Concepts, methods, and findings concerning
nostalgic influences on customer tastes. Journal of Business Research, 37,
27 – 39.
Hyman, I. E., Husband, T. H., & Billings, F. J. (1995). False memories of child-
hood experiences. Applied Cognitive Psychology, 9, 181 – 197.
Hyman, I. E., & Pentland, J. (1996). The role of mental imagery in the creation
of false childhood memories. Journal of Memory and Language, 35, 101 – 117.
Johnson, M. K., Foley, M. A., Suengas, A. F., & Raye, C. L. (1988). Phenomenal
characteristics of memories for perceived and imagined autobiographical
events. Journal of Experimental Psychology: General, 117, 371 – 376.
Krugman, H. E. (1967). The measurement of advertising involvement. Public
Opinion Quarterly, 30, 349 – 356.
Lindsay, D. S., & Read, J. D. (1994). Psychotherapy and memories of childhood
sexual abuse: A cognitive perspective. Applied Cognitive Psychology, 8, 281 –
338.
Loftus, E. F. (1997). Creating false memories. Scientific American, 277(3), 70 –
75.
Loftus, E. F., & Ketcham, K. (1994). The myth of repressed memory. New York:
St. Martin’s Press.
Loftus, E. F., & Loftus, G. R. (1980). On the permanence of stored information
in the human brain. American Psychologist, 35, 409 – 420.
Loftus, E. F., & Pickrell, J. E. (1995). The formulation of false memories. Psy-
chiatric Annals, 25, 720 – 725.
Marconi, J. (1996, October 21). Retro marketing helps brands gain new image.
Marketing News, 30, p. 10.
Neisser, U., & Fivush, R. (1994). The remembered self: Construction and ac-
curacy in the self narrative. Cambridge: Cambridge University Press.
Pezdek, K., Finger, K., & Hodge, D. (1997). Planting false childhood memories:
The role of plausibility. Psychological Science, 8, 437 – 441. short
standard

22 BRAUN, ELLIS, AND LOFTUS Base of RF

MAR WILEJ RIGHT INTERACTIVE

Top of text
Prince, G. W. (2000, March 15). Yesterday, today and tomorrow. Beverage Base of text
Word, pp. 52 – 54.
Puto, C. P., & Wells, W. D. (1983). Informational and transformational adver-
tising: The differential effects of time. In T. K. Kinnear (Ed.), Advances in
consumer research (Vol. 11, pp. 638 – 643).
Roese, N. J. (2000). Counterfactual thinking and marketing: Introduction to
the special issue. Psychology & Marketing, 17, 1.
Schacter, D. L. (1995). Memory distortion. Cambridge, MA: Harvard University
Press.
Schacter, D. L., Koutstaal, W., Johnson, M. K., & Gross, M. S. (1997). False
recollection induced by photographs: A comparison of older and younger
adults. Psychology & Aging, 12, 203 – 215.
Schwarz, N. (1996). Cognition and communication. Mahwah, NJ: Lawrence
Erlbaum.
Sherman, S. J., Cialdini, R. B., Schwartzman, D. F., & Reynolds, K. D. (1985).
Imagining can heighten or lower the perceived likelihood of contracting a
disease: The mediating effect of ease of imagery. Personality & Social Psy-
chology Bulletin, 11, 118 – 127.
Spiegel, D. (1997). Memories: True and false. American Psychologist, 52, 995 –
996.
Sujan, M., Bettman, J. R., & Baumgartner, H. (1993). Influencing consumer
judgments using autobiographical memories: A self-referencing perspective.
Journal of Marketing Research, 30, 422 – 436.
Thompson, C. P., Skowronski, J. J., Larsen, S. F., & Betz, A. (1996). Autobio-
graphical memory: Remembering what and remembering when. Mahwah,
NJ: Lawrence Erlbaum.
Wells, W. D. (1986). Three useful ideas. In R. J. Lutz (Ed.), Advances in Con-
sumer Research (Vol. 13, pp. 9 – 12). Provo, UT: Association for Consumer
Research.

Correspondence regarding this article should be sent to: Kathryn Braun, Mar-
keting Memories娃; P.O. Box 1842, Auburn, AL 36831-1842.

short
standard

MAKE MY MEMORY 23 Base of RF